Resistance to change and preference for variable versus fixed response sequences.
Level 5 - mechanism / opinion, no new human data
Animal laboratory behavioral study with no human data
PubMed 22851789 · doi:10.1901/jeab.2012.98-1
What was done
In Experiment 1, 4 pigeons were trained on a multiple chain schedule with initial links on a variable-interval 20-s schedule and terminal links requiring four responses to left/right keys. In the REPEAT component, only an LRLR sequence was reinforced; in the VARY component, sequences satisfying a variability criterion were reinforced, with reinforcer rates equated across components. Resistance to change was tested with prefeeding, extinction, and response-independent food. In Experiment 2, REPEAT and VARY schedules were terminal links of concurrent chains to assess preference.
What was found
Initial- and terminal-link responding in the VARY component was generally more resistant to prefeeding, extinction, and response-independent food than in the REPEAT component. In Experiment 2, all pigeons biased initial-link responding toward the VARY alternative. The abstract reports these directions but provides no specific numerical values, response rates, or statistical test statistics.
Why it matters
This study demonstrates that behavioral variability is preferred and more resistant to disruption than repetition when reinforcement rates are controlled, presenting challenges for behavioral momentum theory.
Limits
The sample size is very small (4 pigeons), and findings are limited to non-human animal laboratory behavior. The abstract does not report sample size for Experiment 2 or exact quantitative metrics and variance.
Cited by
- supports In psychological operant conditioning studies, animals such as pigeons demonstrate higher persistence and frequency in lever-pressing behavior when provided with variable, unpredictable reward schedules compared to predictable rewards.